-
Notifications
You must be signed in to change notification settings - Fork 16
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
table not found on subsequent "Flame On!" usage #41
Comments
@ruslandoga I am unable to reproduce this issue on the latest code in main (which should be the same as 0.6.0 for the capture portion). Are you still able to reproduce this? If so, are there other logs during/after the first run that might shed some light on the issue? |
Yes, I set up Flame On again and was able to reproduce the issue. The error is the same as before and it keeps repeating. Here's the repro: ruslandoga/plausible#301 $ docker run -d -p 8123:8123 --ulimit nofile=262144:262144 --name plausible_ch clickhouse/clickhouse-server:24.3.3.102-alpine
$ docker run -d -e POSTGRES_PASSWORD=postgres -p 5432:5432 --name plausible_db -v plausible_db:/var/lib/postgresql/data postgres:15-alpine
$ mix do ecto.create, ecto.migrate, run priv/repo/seeds.exs
$ export BASE_URL=https://localhost:8000
$ export SECRET_KEY_BASE=fmjE8dzmLEVUSvH77lM/DUsc868pVbu+RPECVDZz53TjAtB5pRLXY0ND7FQNFh1Q
$ export TOTP_VAULT_KEY=jBr7nua/pCKBHiY/jP1GXhHYeoZ7fyktetOXc56jpJI=
$ export DATABASE_URL=postgres:https://postgres:postgres@localhost:5432/plausible_dev
$ export CLICKHOUSE_DATABASE_URL=https://localhost:8123/plausible_events_db
$ MIX_ENV=ce iex -S mix phx.server Open https://localhost:8000/lv-dashboard/flame_on, start tracing by clicking "Flame On!" $ k6 run test/load/script.js First trace finishes and a flame chart appears. Start the second one by clicking "Flame On!". At this point I start getting errors in the I can try and come up with a minimal repro tomorrow. |
There are no other logs before the error starts appearing. |
@ruslandoga to confirm, this issue is happening in a load testing situation? Meaning the site is being flooded with requests from before the first trace until after the error message? |
@ruslandoga Could you try out the branch mb-41-handle-table-not-started and see if that both resolves the error and still allows repeat traces to be captured? |
Yes. I don't remember it happening outside of load testing.
Sure, I'll try it first thing tomorrow. Thank you! |
I wasn't able to try it out due to some dependency conflicts
I'll try cherry picking the fix into v0.6.0. |
The fix fails with
It should probably be - case ETS.Set.wrap_existing!(__MODULE__) do
+ case ETS.Set.wrap_existing(__MODULE__) do With this change I still get the original error but only once instead of on repeat, and the flame chart appears. Thank you! |
Nice, yes, sorry for that mistake. I have pushed the change to the branch. Can you paste the full error you get now? It shouldn't be the exact same error since the function that raised it is not called anymore. |
You are right, the error should've been different but it was the same. I guess there was some recompilation issue. Either way, I think the original issue is fixed. Thank you again! |
👋
After running "Flame On!" once, I'm getting
on all subsequent requests.
The text was updated successfully, but these errors were encountered: